In existing systems, implementation of multi-tenancy in a database may be accomplished using a technique called striping, wherein each row of a database comprises a column containing a value that indicates the tenant that may access the data in the row. Users may then be associated with the various tenants, making it possible to structure database queries to return only those results from the rows that a user is authorized to access through his association with the tenant indicated. However, in such systems, it is not easy to implement shared access to certain database objects due to their nature, or to change access authorization to database objects from one group of tenants to another.